When a calling terminal inputs a telephone number of a called terminal, a wired telephone network or a mobile communication network requests a call to the called terminal. When the called terminal accepts the call, a call is connected between the calling terminal and the called terminal. If the called terminal is busy, the called terminal is turned off, or a user of the called terminal rejects a call, the call is not connected, and a guidance sound source established by a telecommunication service provider on the side of the called terminal is transmitted to the calling terminal.
For the purpose of efficient management, a company calling a plurality of users in a like manner of a call center uses a calling device for telephoning the called terminal, and when the called terminal responds to it, it is connected to a call center agent so that the same may respond to the called party.
The conventional calling device receives sound source data for notifying a situation of the called terminal for each frame in real time. In this instance, the conventional calling device detects a starting point of the sound source, and analyzes, by comparing patterns and recognizing voice, sound source data after the starting point to determine the situation of the called terminal. Therefore, analysis errors may be influenced according to performance of a starting point detector. And when initial received data are lost, the conventional calling device recognizes a middle point of the actual sound source as the starting point. So it cannot accurately analyze the situation of the called terminal because of asynchronization of a reference sound source and a received sound source. When the called party rejects a call, the sound source provided by the telecommunication service provider comes to each called terminal after an individually established ring back tone. So it is difficult for the conventional device that analyzes the sound source with reference to the starting point to analyze the call rejection situation of the called terminal. Further, the conventional device recognizes the voice (e.g., busy) included in the received data to analyze the situation of the called terminal so there is a large amount of calculation for voice recognition.